Re: Acceptable pressure values

From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Sat May 24 2008 - 09:27:24 CDT

Hi Fabio,
this depends somewhat on what you mean by "acceptable". You're using a
fairly small system over a relatively short time, so having spressure
measurements with large fluctuations and a mean that is not exactly 1
atm . If you're concerned you should calculate the standard deviation of
your pressure measurements (using a method appropriate for correlated
time series) You should at the very least make sure that your cell
volume has stabilized (you may want to look at the pressure averaged
over, say, the last 4 ns of your run to verify that it has leveled out).
Also, if you're using rigidbonds you should be looking at
GPRESSURE/GPRESSAVG instead.

One thing that does sound odd here -- are you sure the mean for PRESSAVG
and PRESSURE that you're quoting are correct? These values should be
nearly identical...

Best,
Peter

Fabio Zegarra wrote:
> Dear NAMD users:
>
> I am simulating a small protein in solution (NPT with around 6000
> water molecules) using langevin dynamics to control the temperature,
> and langevin piston to control the pressure
>
> langevinPiston on
> langevinPistonTarget 1.01325
> langevinPistonPeriod 100.
> langevinPistonDecay 50.
> langevinPistonTemp 300
>
> I have found that the average of PRESSAVG varies between 1.5 and 2;
> while the average of PRESSURE is around -20.
>
> I tried using a wide range of parameters recommended in literature.
> Additionally, I checked the correct minimization and the time of the
> simulation I used is up to 5ns.
>
> I would like to know if these values of PRESSAVG and PRESSURE are
> acceptable.
>
> Thank you in advance,
> Fabio
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:47:50 CST